Prolidase releases both proline and hydroxyproline from imidodipeptides. A body of evidence robustly supports the role of proline in programmed cell death and bioenergetics (Phang J. et al., J. Nutr. 138:2008S, 2008;Phang J et al., Annu. Rev. Nutr. 30:441, 2010), but the other imino acid released by prolidase, hydroxyproline, was also investigated. Hydroxyproline oxidase, an enzyme distinct from proline oxidase, and encoded by a distinct gene, was shown to be a p53-induced gene which generates ROS and activates apoptosis (Cooper, S.K. et al., J. Biol. Chem.283:10485, 2008). These findings further emphasize the importance of prolidase in catalyzing the release of both imino acids from the degradation of extracellular matrix. Previous studies showed that nitric oxide (NO) increases prolidase activity by stimulating its serine/threonine phosphorylation. The effect of NO can be produced either by treatment with an exogenous donor (DETA/NO) or by transfection of cells with iNOS to increase endogenous production of NO. (Surazynski et al., Int. J. Cancer 122:1435, 2008). The effect is mediated through the cGMP-Protein Kinase G pathway rather than the MAP kinase pathway. Since defective wound healing is a prominent finding in patients with inherited prolidase deficiency, we focused on prolidase-dependent mechanisms and showed that prolidase-dependent regulation occurs by least two pathways. First, we showed that the release of proline from collagen degradation increases the rate of collagen synthesis (Surazynski A. et al., Amino Acids 35:731, 2008). Presumably this occurs because substrate proline may be rate limiting for collagen synthesis during wound healing. Another effect of prolidase is mediated through HIF-1alpha. We obtained RKO colorectal cancer cells stably transfected with expression of prolidase 15-20 fold that of vector controls. The prolidase expressing cells (PL) have increased levels of vascular endothelial growth factor (VEGF) in whole cell extracts by western blots and in conditioned media as measured by ELISA. Glucose transporter-1 (Glut-1) is also increased in PL cells. HIF-1 was much more active in PL cells, and the degradation of the oxygen-dependent domain (ODD) was markedly decreased That these mechanisms are due to the catalytic activity of prolidase was shown by the finding that medium proline and hydroxyproline could augment the effect. Furthermore, inhibition of prolidase activity by N-benzyloxycarbonyl-L-proline (Cbz-Pro) markedly decreased the angiogenic signaling with increased prolidase. In conclusion, accompanying the activation of MMP and increased ECM degradation as a source of stress substrates (proline/hydroxyproline), signals are generated to activate angiogenesis to augment the nutrient supply (Surazynski A. et al., Int. J. Cancer 122:1435, 2008). The defective wound healing in patients with prolidase deficiency is of considerable interest since cancer has been considered "a wound that never heals." Thus, we initiated a project to develop prolidase knockout mice. We have obtained ES cells with pepd (prolidase) knocked out and will be injecting them into recipients soon. Importantly, we have initiated a collaboration with Dr. Teresa Gunn of the McLaughlin Research Institute who discovered a 4-bp deletion in the pepd gene coding for prolidase. We have corroborated her findings by showing these mice are deficient in prolidase enzyme activity, and we are now raising a colony in the Frederick mouse facility. After characterizing the impact of this enzymatic defect on proline metabolism, we will test the functional consequences in experiments of wound healing and in carcinogenic models i.e. two-stage skin tumor model.